Key words:
Acrylic resin - biofilm - microbial adhesion - ocular prosthesis
INTRODUCTION
An ocular prosthesis is an option for rehabilitation of anophthalmic patients, being
fabricated principally from acrylic resin. Although the ocular prosthesis can be adopted
adequately in the anophthalmic cavity, in the majority of cases a “dead space” is
observed between the posterior surface and the bottom of the cavity. Lacrimal secretion,
mucous, and stagnant residue in this space constitutes an excellent culture environment
for the growth of bacteria.[1]
One of the most important pathogens of prosthetic infections is the Staphylococcus.[2] Staphylococcus epidermidis can adhere and proliferate on polymer surfaces, especially on lenses and intraocular
prostheses.[3],[4] Staphylococcus aureus lives principally on mucous surfaces and is considered as one of the most versatile
and dangerous human pathogens.[5] Enterococcus faecalis is a natural Gram-positive streptococcal of the intestinal tract, able to cause serious
infections such as endophthalmitis and corneal ulcers.[6],[7] Another microorganism encountered in the ocular region is Candida albicans, which also occurs among the fungal infections of maxillofacial prostheses.[8] The hematogenic dissemination of C. albicans to the eye is associated with intravenous drug abuse, grave debilitation, or immunodepression,
recent surgery (especially gastrointestinal), the use of broad-spectrum antibiotics,
diabetes, and alcohol abuse, among others.[9],[10]
Recently, a large number of reports in respect to the impact of the physical properties
of the materials in the adhesion of microbe,[11]
[12]
[13]
[14]
[15]
[16]
[17] and a particularly strong relation between the bacterial adhesion and surface roughness,
have been highlighted.[14]
[15]
[16]
[17]
[18] The surface roughness of the material is considered a relevant property for the
process of microbial adhesion. This is because polymer surface irregularities, such
as grooves or fissures, promote an increase in the surface area, and depressions that
provide more favorable sites for colonization which protect the microbe against the
forces of shearing.[11],[19] Since the behavior of microbial adhesion on polymers depends on the microorganism
colonizer and the surface roughness of the material, it is important to evaluate the
relation of the surface roughness of acrylic resins for ocular prostheses in the adhesion
of different microorganisms.
Therefore, this study has the objectives of simulating different surface roughnesses
of acrylic resins used in the fabrication of artificial sclera for ocular prostheses,
and to evaluate the interference of roughness on the adhesion and formation of biofilm
of different microorganisms (S. epidermidis, S. aureus, E. faecalis and C. albicans). The null hypotheses of this study were that the surface roughness of acrylic resin
does not differ among the different kinds of surface polishings and that the adhesion
of microorganisms would not be influenced by the roughness of the acrylic resin.
MATERIALS AND METHODS
For the fabrication of the test specimens in acrylic resin, a molded metallic matrix
was used which contained ten circular compartments in its interior, each with dimensions
of 10 mm diameter and 3 mm thickness. The matrix adhered to a rectangular glass slide.
The glass slide and matrix set were then placed in a special muffle for polymerization
in a microwave oven (VIPI STG; VIPI Industria, Pirassununga, Sao Paulo, Brazil), filled
with special plaster Type IV (Durone; Dentsply Ind and Com Ltd, Rio de Janeiro, Brazil).
After the plaster crystallized, another glass slide was positioned over the already
enclosed matrix, a contra muffle was positioned, and special plaster type IV was poured
over the surface of the last glass slide. The muffle was opened after the crystallization,
and the N1 heat-activated acrylic resin (white color) for artificial sclera (Artigos
Odontologicos Classico Ltda., Sao Paulo, Brazil) was proportioned, manipulated, and
inserted in the matrix. After the insertion, the contra muffle was positioned and
carried to a hydraulic press, (VH; Midas Dental Produtos Ltda., Araraquara, Sao Paulo,
Brazil) and placed under a force of 1200 kg/F for 2 min. Subsequently, a 30 min bench
polymerization was performed, followed by a 10 min microwave polymerization. After
the resin polymerization, the muffle was opened, and the specimens were removed. The
specimens were then submitted to polishing for
3 min using metallographic sandpaper of different granulations (Buehler, Illinois,
USA). Blue colloidal silica solution (Buehler, Illinois, USA) with a grain of 1 micrometric
(μm) was used on one group. After polishing, ultrasound cleaning was performed to
remove possible debris.
A total of 432 test specimens were fabricated. Of these, 144 were divided for the
periods of 4, 24, and 48 h of microbial growth and adhesion. Thirty-six test specimens
were inoculated with one of the four microorganisms evaluated. For each microorganism,
the test specimens were distributed randomly in six groups (6 specimens per group
with 3 for each experiment): Group 1200S: polished with 1200-grain sandpaper and the
diamond solution of 1 μm; Group 1200: polished with 1200-grain sandpaper; Group 800:
polished with 800-grain sandpaper; Group 400: polished with 400-grain sandpaper; Group
120: polished with 120-grain sandpaper; and Group unpolished. After the surface polishing,
the test specimens were sterilized with ethylene oxide.
Strains of S. epidermidis (ATCC 35984); S. aureus (ATCC29213); E. faecalis (ATCC 29212) and C. albicans (ATCC 90028) were used. All strains were donated by the Oswaldo Cruz Foundation-
FIOCRUZ, Rio de Janeiro, Brazil. The microorganisms were maintained at -70°C in a
solution containing 25% glycerol, seeded in plates containing culture medium adequate
for each microorganism, and incubated aerobiotically at 37°C for 24 h. The culture
mediums used were as follows: Mannitol Salt Agar (Difco, Kansas City, MO, USA) for
S. epidermidis and S. aureus; m-Enterococcus Agar (Difco, Kansas City, MO, USA), for E. faecalis, and Sabouraud Dextrose Agar (Difco, Kansas City, MO, USA) for C. albicans.
Initially, growth curves were performed for each microorganism with the purpose of
identifying the number of hours necessary for each to achieve its greatest microbial
multiplication phase (log phase), which was determined by the optic density values.
The cultures were then seeded in BHI broth (Difco) or Sabouraud Dextrose and maintained
at 37°C in an aerobiotic incubator for 24 h. The microorganisms were adjusted halfway
through the logarithmic phase, being diluted ×10, ×100, or ×100 in BHI broth with
×2 concentration, depending on each microorganism, to obtain 107 microorganisms/ml.
After growth and obtainment of 107 microorganisms/ml, 1 ml of medium containing the microorganism that was used was
placed in contact with the acrylic resin test specimens positioned in 24-well microtiter
plates. Microbial suspensions were incubated in wells without the test specimen for
negative control. The positive control of this study was the BHI broth or sabouraud
dextrose for the tests with C. albicans without the microorganism. The wells were incubated at 37°C in an aerobiotic incubator
for 4, 24, and 48 h. After each period, the nonadhering microorganisms were removed
from the test specimens by way of a wash in 1 ml of saline solution. Subsequently,
each test specimen was inserted in a test tube containing 1 ml of saline solution.
The tubes underwent an ultrasound (USC 700; UNIQUE Ultrasonic Cleaner, Sao Paulo,
Brazil) cleaning bath at 50 kHz, 150 W, for 20 min, and agitation (Vortex QL- 901;
Biomixer, Curitiba, Parana, Brazil) for 1 min, to promote detachment of adhered microorganisms.
Subsequently, a series of seven dilutions were performed, transferring 10 ul of this
solution to 90 ul of saline solution. Dilutions 4 and 7 were plated in petri plates
containing an adequate medium for each microorganism evaluated, and incubated for
37°C in an aerobiotic incubator for 24 h. After the incubation period, the counting
of the colony-forming units (CFU/mL) was done. The tests were performed in duplicate
on two independent days.
The surface roughness of all test specimens was determined using a profilometer (Dektak
d-150; Veeco, Plainview, New York, USA). One test specimen was positioned individually
in the center of the equipment, and the profilometer measuring tip was focused on
its surface. The Ra (surface roughness arithmetic mean) values were measured using
a cut-off of 500 μm in a 12s time constant. Three readings were performed on each
surface, and the mean was calculated. The original values were given in Angstrom (Å),
and were then transformed to the μm scale.
A test specimen from each sandpaper polish group was analyzed using atomic force microscopy
(AFM; Veeco Metrology Inc., Santa Barbara, CA, USA). The images obtained were transported
from the microscope to a computer. Subsequently, they were sent to the NanoScope Analysis
program (2004; Veeco Instruments Inc., Santa Barbara, CA, USA) and submitted to filters
(“lowpass” and “medium”). All three-dimensional (3D) images were standardized in the
minimum height scale of -100 nm and a maximum of 100 nm (z-axis) for later qualitative
comparison among the groups.
The surface roughness value and test specimen microorganism count for each group,
as well as the time of incubation and type of microorganism, were submitted to the
normal curve adherence test to determine if they provided a normal distribution. A
normal distribution for roughness values was found, consequently performing a one-way
ANOVA and parametric Tukey test with a 5% (P ≤ 0.05) level of significance. A nonnormal distribution for CFU data was also found
and hence, appropriate nonparametric statistic tests were applied to compare the measurements
through the Kruskal-Wallis and Mann-Whitney tests.
RESULTS
Using the one-way ANOVA, it was observed that there was a significant statistical
difference comparing the Ra roughness values obtained by the different polishings
(F = 1361.651; P ≤ 0.001), with the 120 and 400 groups differentiated statistically between themselves
and in relation to all the other groups, with Group 120 presenting the greatest Ra
roughness value, followed by Group 400. The nonpolished group did not differ statistically
from Group 800, 1200, or 1200S [Table 1].
Table 1:
Mean values (standard deviation) of Ra roughness of acrylic resin, according to polishing
sandpaper used
|
Acrylic resin
|
|
Groups
|
Ra roughness
|
|
Means followed by the same capital letter in column do not differ at the 5% significance
level (P<0.05; Tukey)
|
|
Unpolished
|
0.05 (0.04) C
|
|
120
|
1.72 (0.36) A
|
|
400
|
0.52 (0.12) B
|
|
800
|
0.03 (0.01) C
|
|
1200
|
0.02 (0.01) C
|
|
1200s
|
0.03 (0.01) C
|
The qualitative analysis regarding the surface smoothness of 3D images of the study
reveals apparent differences in the images, as to the formation of irregularities
along the extensions evaluated. Groups 120 and 400 present images that reveal irregularities
with more significant cracks and orifices. These include the interior of a crack with
the formation of a deeper “valley” without the definition of the highest point above
the “peaks,” as compared to those that still maintain the characteristic of a polished
surface (800, 1200, and 1200S, and nonpolished), although the nonpolished group presented
small irregularities on its surface [Figure 1] 1.1-1.6].
Figure 1: Representative atomic force microscopy image. 1.1 represents the Group 120 and 1.2
represents the Group 400 at the both images irregular surfaces are evident. 1.3 represents
the Group 800; 1.4 represents the Group 1200 and 1.5 represents the Group 1200s at
this images the characteristics of polished surfaces were maintained. 1.6 represents
the Group Unpolished, which presents smooth surface with the presence of some peaks
and valleys
[Figures 2] and [3] show the microbial growth values for the bacterial species. It can be observed that
for S. epidermidis, all of the groups differentiate from the control group at 4 and 48 h. Group 1200S
differentiated from the control in all time periods and had the smallest bacterial
growth values after 24 and 48 h of growth [Figure 2].
Figure 2: Mean (standard deviation) of colony-forming unit/ml count (in logarithmic scale)
of Staphylococcus epidermidis for each granulation group. a: Different lowercase letters show statistical difference
between each group, according to the Mann-Whitney tests, considering P ≤ 0.05
Figure 3: Mean (standard deviation) of colony forming unit/ml count (in logarithmic scale)
of Enterococcus faecalis for each granulation group. a: Different lowercase letters show statistical difference
between each group, according to the Mann-Whitney tests, considering P ≤ 0.05
[Figure 4] shows that for S. aureus, only Group 1200S presented the smallest bacterial growth values, with the statistical
difference about the control.
Figure 4: Mean (standard deviation) of colony-forming unit/ml count (in logarithmic scale)
of Staphylococcusaureus for each granulation group. a: Different lowercase letters show statistical difference
between each group, according to the Mann-Whitney tests, considering P ≤ 0.05
For E. faecalis, it was observed that the initial adhesion probably took more than 4 h to occur since
there was no microorganism growth observed in this period. In addition, except Group
400 at 48 h, all the materials differentiated from the control, but not among themselves
[Figure 3].
Adhesion of C. albicans was not observed on the acrylic resin surfaces evaluated.
DISCUSSION
The first hypothesis of this study that the different acrylic resin surface roughnesses
do not interfere with the growth of microorganisms was confirmed by the fact that
there were no significant differences among the different polish groups. The second
hypothesis that the different sandpaper grains for polishing do not promote significant
differences in the surface roughness values was partially accepted since there was
a significant statistical difference for E. faecalis in the growth of microorganisms in the group polished with 400 grain sandpaper. The
bacterial growth model used in this study simulated in vitro conditions of static biofilm growth that is encountered on the contact surface of
an ocular prosthesis with the conjunctive membrane tissue.
The initial microorganism adhesion to the surface of the materials is the key requirement
for the colonization of the material.[19],[20] When a bacterium adheres and proliferates on the surface of a material, it produces
extracellular polymer substances and forms the biofilm, which covers and protects
against the immune systems and antimicrobial agents.[18] During the process of adhesion, the bacteria adheres firmly to the surface of the
material through physicochemical interactions which include the hydrophobicity and
charge of the cell surface, as well as the chemical composition and the surface roughness
of the material.[19],[21]
Several published studies showed that in vivo bacterial adhesion is determined primarily by a Ra surface roughness >0.2 μm.[18],[22]
[23]
[24]
[25] However, this study observed that, in general, even the groups with lower mean Ra
values (0.02, 0.03, and 0.05) demonstrated bacterial growth [Table 1] and [Figures 2]-[4]. This data corroborate with Yoda et al.,[18] which evaluated the adhesion of microorganisms on surfaces of different biomaterials
and obtained a result which showed that even surfaces with roughness levels below
30 nm (0.03 μm) could promote bacterial adhesion. In addition, Lee et al.
[26] observed bacterial growth with roughness levels below 0.2 μm on composite resin
(Ra = 0.179), titanium (Ra = 0.059), and zirconia (Ra = 0.064) surfaces.
These affirmations indicate that there is no consensus with regard to the minimum
roughness level for microbial adhesion and could differ according to the material
used and the capacity of the microorganism to adhere to different surfaces. Thus,
bacterial adhesion is a multifactorial phenomenon, and surface roughness is not the
only influential characteristic in the process.[18] In this way, other factors are also connected to the potential microbial adhesion
to the acrylic resin surface, such as hydrophobicity, electrostatic interactions,
and surface energy.[27] The chemical composition of the material, the adhesion capacity level, and the size
and mode of microorganism division also play a role. These factors could explain why
E. faecalis took more time to adhere to the acrylic resin surface [Figure 3] compared with the staphylococcus since the later presented a characteristic of early
adhesion.[19]
Published studies report that microorganisms appear to have a preference for adhesion
on rougher surfaces with scratches and grooves.[19] However, the current study observed that except Group 1200S, which received polishing
with 1200-grain sandpaper and silicone-based diamond solution, the other groups did
not differ among themselves as far as the bacterial growth values [Figures 2]-[4]. This data corroborate with Taha et al.,[23] which evaluated microorganism adhesion, including S. aureus and C. albicans on 3 types of orthodontic wire with different roughnesses, and did not encounter
a significant difference among the groups.
In agreement with the results of the current study, some authors[28]
[29]
[30]
[31] reported that a linear relation between bacterial adhesion and surface roughness
is not always observed. A small increase in roughness could lead to a significant
increase in bacterial adhesion, while a large increase in roughness might not have
a significant effect on adhesion.[19] Previous published studies[24],[25] reported that small variations of surface roughness do not present a significant
effect on bacterial adhesion, which could justify the absence of a significant relationship
between microbial adhesion and the roughness observed in this study.[23]
Although the AFM images show a certain irregularity in the surface of the nonpolished
group [Figure 1f], the Ra mean value of this group was low and similar to the 800, 1200, and 1200S
Groups [Table 1]. Still, the nonpolished group did not differ statistically for the polished group
about bacterial growth [Figures 2]-[4]. This data could have resulted from the compression of the acrylic resin against
a glass slide at the moment of fabrication of the test specimens with polish, which
could have resulted in a smooth surface similar to the groups that received refined
polishing, and thus, similar roughness and bacterial growth values were observed.[22],[32]
It was observed that the group with more refined polishing (1200S) was statistically
different from the control for all periods and bacteria evaluated, principally for
S. epidermidis and S. aureus. In general, this group presented smaller values of bacterial growth, although it
can be observed in the AFM images[33] and the Ra roughness values that there was no difference about the 800, 1200, and
nonpolished groups [Table 1] and [Figures 1]-[4]. This result could have occurred due to the silicone ion (Si) implantation present
in the diamond solution used in the polishing of this group, which made the surface
less susceptible, affecting the adhesion process. The Si ion can reduce the surface
energy and the contact angle, influencing the wettability and hampering the adhesion
of microorganisms. The result obtained is in agreement with the studies of Zhao, in
which it was observed that there was a reduction in microorganism adhesion in the
presence of SiF3 in stainless steel.[32] In addition, Rashid et al.
[34] reported that the use of the diamond solution for polishing favored the prevention
of microorganism accumulation on the surface of the materials.
An interesting fact observed in the results of the current study, which differs from
the majority of studies encountered[27],[35],[36] was that the C. albicans did not adhere to the surfaces of the test specimens, independent of the polishing
realized.[37] The low indices of C. albicans adhesion may be associated with the morphology of the fungus. This may be because
C. albicans fungus is large (4-6 μm)[35] and possess long filaments in their shapes, limiting adhesion to the narrow recess
of surface cracks, making it less stable.[34] It is known that the presence of bacteria facilitates the adhesion of Candida fungus to acrylic resin, principally through the production of extracellular polymers
as well as through an increase in acidity, which creates favorable environmental conditions
for the growth of fungus.[36],[38] In this study, however, isolated biofilms of each microorganism were evaluated,
and in this way, the C. albicans strains could not count on the presence of other microorganisms to aid in their growth.
Although the results of this study revealed a weak relation between surface roughness
and microbial adhesion, the stages of polishing during the fabrication of ocular prostheses
should not be overlooked, since bacterial adhesion is not determined only by this
isolated property. In addition, roughness should never be considered alone about microbial
adhesion. Other factors are also associated with this property, such as comfort and
satisfactory esthetics of the patient.
Diverse limitations should be considered to interpret the results. The pathogenicity
of prosthetic devices is a complex process involving interactions between the pathogen,
material, and the host. An in vitro study cannot count on the defense of the host and other factors such as fluctuation
of temperature conditions and nutrition.[18]
Conclusions
Thus, considering the limitations described, and even though, the low values of roughness
obtained through different polishings could have been important in preventing the
adhesion of C. albicans, it can be concluded that the small variations in acrylic resin surface roughness
did not produce differences in adhesion or formation of biofilms of the bacteria evaluated.
This suggests that roughness is not the only property that should be considered when
evaluating microbial adhesion. Silica appears to interfere with microorganism adhesion
since microbial growth decreases significantly compared with other groups and the
control group when silica solution is associated with polishing.
Financial support and sponsorship
Nil.
